Image cropping with lossless resolution for generating enhanced image databases

Abstract

A method including selecting, in a server, a first image portion from an image is provided. The method also includes identifying one or more known similar images associated with the first image portion, and determining a first score for enhancing the first image portion based on the known similar image(s). The method includes increasing a pixel resolution in the first image portion according to the scale to form an enhanced image portion. The method also includes identifying a synthetic value for the enhanced image portion and storing the enhanced image portion in a database when the synthetic value is below a tolerance value.

What is claimed is: 
     
       1. A computer-implemented method, comprising:
 selecting, in a server, a first image portion from an image; 
 identifying one or more known similar images associated with the first image portion; 
 determining a first score for enhancing the first image portion based on the one or more known similar images; 
 increasing a pixel resolution in the first image portion according to a scale to form an enhanced image portion; 
 identifying a synthetic value for the enhanced image portion; and 
 storing the enhanced image portion in a database when the synthetic value is below a tolerance value. 
 
     
     
       2. The computer-implemented method of  claim 1 , further comprising receiving the image from a user via a client device communicatively coupled to the server. 
     
     
       3. The computer-implemented method of  claim 1 , further comprising retrieving the image from the database upon a user request. 
     
     
       4. The computer-implemented method of  claim 1 , wherein selecting the first image portion from the image comprises parsing the image according to a salient feature in a classification space, and verifying that the enhanced image portion contains the salient feature. 
     
     
       5. The computer-implemented method of  claim 1 , further comprising selecting a second image portion from the image, determining a second score for enhancing the second image portion based on a second similar image from the database, and forming a second enhanced image portion based on the second image portion when the second score is greater than the first score. 
     
     
       6. The computer-implemented method of  claim 1 , wherein identifying the one or more known similar images comprises receiving the one or more known similar images from a user via a client device communicatively coupled to the server. 
     
     
       7. The computer-implemented method of  claim 1 , further comprising adjusting the scale for the first image portion using a Generative Adversarial Network (GAN) to increase a pixel resolution and obtain the enhanced image portion that resembles the one or more known similar images and has the synthetic value below the tolerance value. 
     
     
       8. The computer-implemented method of  claim 1 , wherein increasing a pixel resolution in the first image portion comprises:
 increasing a number of pixel rows or a number of pixel columns according to an aspect ratio of the one or more known similar images in the enhanced image portion, and 
 selecting a new pixel value to enhance a salient feature of the image in a classification space. 
 
     
     
       9. The computer-implemented method of  claim 1 , wherein increasing a pixel resolution in the first image portion comprises splitting at least one pixel in the first image portion into two or more enhanced pixels, each of the enhanced pixels having a different pixel value based on a salient feature of the image in a classification space. 
     
     
       10. The computer-implemented method of  claim 1 , further comprising modifying a crop factor in the first image portion when the first score is less than a pre-selected threshold. 
     
     
       11. A system comprising:
 one or more processors; and 
 a memory coupled to the one or more processors, the memory including instructions that, when executed by the one or more processors, cause the one or more processors to: 
 select, in a server, a first image portion from an image; 
 identify one or more known similar images to the first image portion; 
 determine a first score for enhancing the first image portion based on the one or more known similar images; 
 increase a pixel resolution in the first image portion according to a scale to form an enhanced image portion; 
 identify a synthetic value for the enhanced image portion; and 
 store the enhanced image portion in a database when the synthetic value is below a tolerance value. 
 
     
     
       12. The system of  claim 11 , wherein the one or more processors further execute the instructions to select a second image portion from the image to determine a second score for enhancing the second image portion based on a second similar image from the database, and to form a second enhanced image portion based on the second image portion when the second score is greater than the first score. 
     
     
       13. The system of  claim 11 , wherein to select the first image portion from the image the one or more processors execute the instructions to parse the image according to a salient feature in a classification space, and to verify that the enhanced image portion contains the salient feature. 
     
     
       14. The system of  claim 11 , wherein to identify the one or more known similar images the one or more processors receive the one or more known similar images from a user via a client device communicatively coupled to the server.